English

Improved Bounds for Randomly Sampling Colorings via Linear Programming

Data Structures and Algorithms 2018-11-01 v1 Discrete Mathematics Mathematical Physics math.MP Probability

Abstract

A well-known conjecture in computer science and statistical physics is that Glauber dynamics on the set of kk-colorings of a graph GG on nn vertices with maximum degree Δ\Delta is rapidly mixing for kΔ+2k\ge\Delta+2. In FOCS 1999, Vigoda showed that the flip dynamics (and therefore also Glauber dynamics) is rapidly mixing for any k>116Δk>\frac{11}{6}\Delta. It turns out that there is a natural barrier at 116\frac{11}{6}, below which there is no one-step coupling that is contractive with respect to the Hamming metric, even for the flip dynamics. We use linear programming and duality arguments to fully characterize the obstructions to going beyond 116\frac{11}{6}. These extremal configurations turn out to be quite brittle, and in this paper we use this to give two proofs that the Glauber dynamics is rapidly mixing for any k(116ϵ0)Δk\ge\left(\frac{11}{6} - \epsilon_0\right)\Delta for some absolute constant ϵ0>0\epsilon_0>0. This is the first improvement to Vigoda's result that holds for general graphs. Our first approach analyzes a variable-length coupling in which these configurations break apart with high probability before the coupling terminates, and our other approach analyzes a one-step path coupling with a new metric that counts the extremal configurations. Additionally, our results extend to list coloring, a widely studied generalization of coloring, where the previously best known results required k>2Δk > 2 \Delta.

Keywords

Cite

@article{arxiv.1810.12980,
  title  = {Improved Bounds for Randomly Sampling Colorings via Linear Programming},
  author = {Sitan Chen and Michelle Delcourt and Ankur Moitra and Guillem Perarnau and Luke Postle},
  journal= {arXiv preprint arXiv:1810.12980},
  year   = {2018}
}

Comments

This is a merger of arxiv:1804.04025 and arxiv:1804.03156. Preliminary version accepted to SODA 2019

R2 v1 2026-06-23T04:58:19.267Z